The purpose of this project is the application of information theory to basic and clinical research on the relationships between genomic DNA, spliced mRNA and resulting protein sequences. We have collaborated with Dr. T. Schneider of NCI, et al., to perform DNA splice site analyses. This includes collaborative development of processing algorithms for the information content of macromolecular sequences. It also involves communication of data, processing methods, and results among researchers in diverse fields.[unreadable] [unreadable] Splice site analyses of some splice variants in the ASPM gene were done for Dr. Vladimir Larionov (NCI/CCR) with Dr. Barry Zeeberg (NCI/CCR). This is a gene with two well-accepted alleles, and possibly a fairly large number of others that have not been reported. Analysis showed that splicing events almost always occurred at sites predicted to be strong to moderate by our local analysis. This gene is rich in potential sites, so determining the sites of likely splicing events is quite within our capabilities. However, choosing most likely sets of exons and calculating probabilities of alleles has been much more difficult. We are trying to determine the probabilities using slightly more global constraints, as discussed in the companion report, "Information Analysis of DNA, RNA, and Protein Sequences."[unreadable] [unreadable] A manuscript describing our work on the Neufibromatosis, NF2, gene mutations was substantially revised.[unreadable] [unreadable] We, in consultation with Drs. Schneider and Rogan, agreed to provide analyses of the information theoretic splice site strengths for several hundred mutations associated with a comprehensive resource of disease-causing mutations, associated transcripts, and the results of several methods used to predict their effects. The overall collection is accessible through a World Wide Web site that is being coordinated by Dr. Igor Vorechovsky, of the University of Southampton School of Medicine, in the United Kingdom. There are certain differences, which we will work to resolve, between the way that we use our method and the way it is used in comparison with the consensus sequence, maximum likelihood, Markov, and other models in this resource.